12 October 2023 in IEEE Transactions on Automation Science and Engineering
In this paper, we introduce the novel state-of-the-art Dual-attention Transformer and Discriminative Flow (DADF) framework for visual anomaly detection. Based on only normal knowledge, visual anomaly detection has wide applications in industrial scenarios and has attracted significant attention. However, most existing methods fail to meet the requirements of logic defect detection under complex semantic conditions. In contrast, the proposed DADF presents a new paradigm: it firstly leverages a pre-trained network to acquire multi-scale prior embeddings, followed by the development of a vision Transformer with dual attention mechanisms, namely self-attention and memorial-attention, to achieve global-local two-level reconstruction for prior embeddings with the sequential and normality association. Additionally, we propose using normalizing flow to establish discriminative likelihood for the joint distribution of prior and reconstructions at each scale. The experimental results validate the effectiveness of the proposed DADF approach, as evidenced by the impressive performance metrics obtained across various benchmarks, especially for logic defects with complex semantics. Specifically, DADF achieves image-level and pixel-level AUROC scores of 98.3 and 98.4, respectively, on the Mvtec AD benchmark, and an image-level AUROC score of 83.7 and a pixel sPRO score of 67.4 on the Mvtec LOCO AD benchmark. Additionally, we applied DADF to a real-world Printed Circuit Board (PCB) industrial defect inspection task, further demonstrating its efficacy in practical scenarios. The source code of DADF is available at https://github.com/hmyao22/DADF. Note to Practitioners —Most of the current industrial visual inspection techniques can only detect structural defects under uncomplicated semantic settings. Detecting anomalies in products featuring intricate components and logical defects with high-level semantics remains a considerable challenge. The presented DADF is a robust model that can effectively identify defects in products with complex components, such as Printed Circuit Boards (PCBs). Furthermore, it can also accurately detect both structural and logical defects, which is of significant importance for practical industrial applications.
13 September 2023 in IEEE Transactions on Circuits and Systems for Video Technology
This paper presents a novel framework, named Global-Local Correspondence Framework (GLCF), for visual anomaly detection with logical constraints. Visual anomaly detection has become an active research area in various real-world applications, such as industrial anomaly detection and medical disease diagnosis. However, most existing methods focus on identifying local structural degeneration anomalies and often fail to detect high-level functional anomalies that involve logical constraints. To address this issue, we propose a two-branch approach that consists of a local branch for detecting structural anomalies and a global branch for detecting logical anomalies. To facilitate local-global feature correspondence, we introduce a novel semantic bottleneck enabled by the visual Transformer. Moreover, we develop feature estimation networks for each branch separately to detect anomalies. Our proposed framework is validated using various benchmarks, including industrial datasets, Mvtec AD, Mvtec Loco AD, the logical dataset DigitAnatomy, and the newly proposed Mvtec AAD dataset. Experimental results show that our method outperforms existing methods, particularly in detecting logical anomalies.
18 December 2023 in IEEE Transactions on Instrumentation and Measurement
In recent years, the field of industrial visual anomaly detection (VAD) has attracted significant attention in the context of advanced smart manufacturing systems. However, several limitations remain unresolved in existing approaches. While these methods can achieve satisfactory performance when training separate models for different categories, their scalability and performance suffer when faced with the challenge of simultaneous training for multiple categories. Reconstruction-based methods generally suffer from the identical mapping problem. To address these limitations, this study introduces the partial semantic aggregation vision transformer (PSA-VT), a scalable framework for industrial visual anomaly detection (VAD) that enables simultaneous multicategory anomaly detection using a single model. Our proposed PSA-VT framework adopts a hybrid design strategy. First, a pretrained convolutional neural network (CNN) is employed to extract multiscale discriminative local representation. Subsequently, the PSA-VT is introduced to perform representation reconstruction through long-range global semantic aggregation. Finally, the anomalous properties can be estimated by evaluating the reconstruction error of the representations. We conducted extensive experiments using the Mvtec AD industrial anomaly detection dataset, as well as the semantic anomaly detection datasets. The experimental results demonstrate that our method achieves state-of-the-art (SOTA) performance by capturing high-level semantics. Notably, PSA-VT surpasses other methods for the one-model-15-category anomaly detection tasks on the Mvtec AD dataset. Furthermore, we applied incremental learning techniques to enable the rapid deployment of PSA-VT in a real industrial scenario.
29 June 2021 in IEEE Transactions on Instrumentation and Measurement
The unclear and unpredictable fracture surface orientations of cortical bones have restricted the deep understanding of fracture mechanisms. The multidirectional loads aggravate the complexity and difficulty in the measurement of the fracture surface orientations. This article focused on novel testing instrument and measurement method of the fracture surface orientations of cortical bones. The controllable external loads were realized by a self-developed in situ tester under approximate in vivo environment and various sampling orientations of bone specimens. A normal vector measurement method was proposed to obtain the normal vector of fracture surfaces under various sampling angles. Quantitative variation of fracture surface orientations along with the direction of external loads was obtained. As the sampling angle gradually increased in a range from 0° to 60°, the normal vector angles exhibited a “slow first then fast later” monotonous increasing tendency in a range from 69.18° ± 9.40° to 84.48° ± 9.87°. On the basis of proposed complete fitting method and eliminating abnormal point method, the measurement error of normal vector angle exhibited a slight fluctuation in a range from 0.22% to 1.83%, dependent on the straightness error of intersection line and flatness error of fracture surface. The approximatively invariable propagation distance along the osteon/interstitial lamellae interface and the “bending ductility effect” induced longer propagation distance inside the osteons resulted in the increased normal vector angles of the fracture surfaces when the external loads gradually deviated from the long axis of bones.
18 May 2023 in IEEE Transactions on Instrumentation and Measurement
Non-dispersive infrared (NDIR) sensors have been widely used in measuring gas concentration in process industry. However, the inhomogeneous gas diffusion in the gas chamber of NDIR sensor increases the response time and causes measurement error. In this case, the fast-and-accurate measurement of gas concentration is of great importance in avoiding accidents. This article focuses on constructing a novel gas diffusion model among normalization concentration (N-concentration), inlet flow velocity, and diffusion time. Firstly, the structure model of gas chamber is constructed based on Fluent 15.0 and the turbulence intensity is calculated based on dynamic viscosity, inlet flow velocity, concentration, temperature and pressure. The simulation model is constructed based on the assumption of simultaneous reaction and mass transfer. The initial result shows that the N-concentration along the optical path is mainly influenced by inlet flow velocity and diffusion time, while the temperature, pressure and inlet concentration have less influence on gas diffusion. According to the velocity distribution map, the critical zone of inlet flow velocity is defined to divide the gas diffusion state into different zones. Then, a novel Multi-interval exponential diffusion model is proposed to calculate the N-concentration and the real-time measurement of gas concentration can be achieved based on inlet flow velocity, diffusion time and the pre-defined critical zone of inlet flow velocity. The proposed model is suitable for the gas concentration correction of different gases. Finally, the validation indexes include the coefficient of determination (R 2 ), Mean Squared Error (MSE) and standard uncertainty shows the accuracy of the proposed model can be obviously increased.
22 November 2025 in Knowledge-Based Systems